Vehicle Sensor Roll Angle Estimation from Stationary Object Reflections
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driver assistance systems in vehicles face inaccuracies in object positioning due to sensor misalignment, leading to incorrect system responses, as current calibration methods rely on specific conditions like parallel lanes or guide rails, limiting their applicability and robustness.
Innovation Solution
A method to estimate the sensor mounting roll angle using two-dimensional histograms of signal strength over horizontal and vertical angles from stationary object detections, without requiring specific calibration arrangements, by filtering and accumulating detections based on vehicle velocity and kinematic data, and evaluating a linear relationship to determine the roll angle.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If calibration methods rely on specific conditions like parallel lanes or guide rails, then measurement precision can be improved, but adaptability deteriorates
Solution Approach 1:
The system performs self-calibration by automatically detecting stationary objects in the environment and using their reflection characteristics to determine sensor roll angle, eliminating the need for external calibration infrastructure like guide rails or parallel lanes
Solution Approach 2:
The method changes the calibration approach from relying on fixed environmental geometries to using variable environmental objects (stationary objects with different orientations), allowing calibration to work in diverse driving scenarios including urban areas without specific road markings
2Ease of manufacture
If sensor mounting deviates from nominal mounting angle, then ease of installation is improved, but measurement precision deteriorates
Solution Approach 1:
The system measures the actual roll angle of the sensor by analyzing the distribution of reflection points from stationary objects and provides feedback to correct the mounting deviation, allowing the sensor to function accurately even when not perfectly aligned during installation
Solution Approach 2:
Instead of relying on precise mechanical mounting, the patent uses a computational approach where software algorithms analyze reflection patterns to detect and correct mounting errors, replacing the need for precise mechanical alignment
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Provides robust and accurate estimation of sensor alignment, enabling precise object detection and system calibration, applicable in various driving scenarios, including urban areas, without reliance on road clutter signals.
Implementation Method 1
a radar sensor or a lidar sensor is used for monitoring the surrounding traffic
Implementation Method 2
the core of the invention is that the roll angle of the sensor is estimated based on the reflection angle from stationary objects
Implementation Method 3
a radar sensor or a lidar sensor is used for monitoring the surrounding traffic
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
A method of detecting a sensor mounting roll angle (ϕ) of an environment detection sensor (10) mounted to a vehicle (12), the method comprising: obtaining (14) detection data of a plurality of detections from the sensor (10), the detection data including signal strength, velocity information, and angle information, the angle information comprising horizontal angle information and vertical angle information; selecting (22), from the detections, detections of stationary objects, based on velocity information of the detections and based on velocity information of the vehicle (12); entering (28) the selected detections into a two-dimensional histogram of signal strength over horizontal angle and vertical angle; evaluating (40) the two-dimensional histogram to determine an estimate of the sensor mounting roll angle (ϕ).